Pituitary atypical adenomas have been referred to have aggressive behavior, such as invasiveness and high rate of recurrence (1,2). From a clinical point of view, preoperative discrimination between atypical adenomas and typical adenomas is important. The purpose of this study is to evaluate the utility of histogram analysis of apparent diffusion coefficient (ADC) value for distinguishing pituitary atypical adenomas from typical adenomas (3,4,5).Between 2012 and 2015 , eighteen patients with pituitary macroadenoma underwent diffusion-weighted imaging (TR/TE, 2500/66ms; b values, 0 and 1000 sec/mm²) and corresponding ADC maps were obtained. Among them, four cases with extensive cystic or hemorrhagic changes were excluded from the analysis. In all cases, pathological confirmations including immunohistochemical examination were obtained from the surgically resected specimen. The ADC maps were reviewed using Ziostation2 (Ziosoft Inc., Tokyo, Japan), and placed a 3D volume-of interest (VOI) on the tumor superimposing contrast enhanced MR images (Figure 1). All ADC values were within in the VOI were used to compute the average ADC within the tumor. The ADC values were binned to construct the histogram. Using the histogram mean, mimimum,maximum, quartile, skewness, kurtosis, and percentile ( 5th,10th,25th,50th,75th, 90th,95th ) of ADC of the entire tumor were computed. Histogram parameters were then compared between atypical adenomas (n=3) and typical adenomas (n=11). Skewness and kurtosis of ADC histogram were significantly lower for atypical adenoma as compared with typical adenoma ( 0.118 ± 0.493 vs 1.419 ± 0.556×10^-6 mm²/sec, P=0.011; 0.720 ± 0.948 vs 3.479 ± 1.353×10^-6 mm²/sec, P=0.011, respectively). No significant differences of ADC histogram were found at other parameters (Table 1). Figure 1 show the representative cases.

The present study showed that histogram parameters including skeweness and kurtosis had additional value for distinguishing pituitary atypical adenomas from typical adenomas. Skewness and kurtosis describe the degree asymmetry and sharpness of a histogram, respectively. We hypothesize that the diffusion pattern of the water molecules is more heterogenous and more non-Gaussian in atypical adenomas than in typical adenomas.

In conclusion, ADC histogram analysis on the basis of entire tumor volume can be useful in distinguishing atypical adenomas from typical adenomas.